[Uim] Proposal for SigScheme

[YamaKen]

Hi Kazuki,

At Wed, 27 Jul 2005 15:38:28 +0900,
mover at hct.zaq.ne.jp wrote:
> Now SigScheme supports proper tail-recursion r1042.

Thank you for the active development effort.

I've written a testsuite for the feature as r1145. The proper
tail recursion is basically works fine. But current SigScheme
implementaion does not have at least two R5RS compliant behavior
about the feature. The 'and' and 'or' conditionals is not
properly tail-recursive (tests for them are provided by the
file).

See "3.5 Proper tail recursion" of R5RS and please fix it.

As this result indicates, well enough testsuites are required to
ensure accurate R5RS compliance. Although I wrote this test from
scratch, writing all other tests from scratch will cost you
non-creative heavy job.

So I recommend that importing the testsuites written for other
R5RS implementations. Searching for them will be efficient
investment to stabilize SigScheme. Please consider it.

> Then I'll thinking about implementing the multiple value.

It will make SigScheme feature-frozen for uim 0.6. I'm looking
forward to SigScheme-based uim with great hope.

> > mover at hct.zaq.ne.jp wrote:
> > > > I think that implementing the multiple values and proper
> > > > tail recursion features to port the widely-used SLIB
> > > > library to SigScheme should take precedence to get SRFI
> > > > features as first step.
> > >
> > > Now, here's simple question.
> > > What "proper" tail recursion means? You mean tail recursion
> > > optimization?
> >
> > Yes. It's the useful word to indicate the concept.

-------------------------------
YamaKen  yamaken at bp.iij4u.or.jp